Embed Size (px)
Citation preview
Chapter 6
Pelagic Ecosystems
CHAPTER SUMMARY
Away from coastal boundaries and above the seabed, the pelagic environment
encompasses the entire watercolumn of the seas and oceans. The pelagic environment
extends from the tropics to the polar regions, and from the sea surface to the abyssal
depths, and is a highly heterogeneous and dynamic three-dimensional habitat. The
pelagic is home to some of the most revered and reviled marine inhabitants, but great
whales and jellyfish alike are subject to the consequences of pelagic ecosystem vari-
ability. Physical processes in the pelagic exert major control on biological activity, and
lead to substantial geographic variability in production. Knowledge of biophysical
interactions is essential for understanding ecological patterns and processes in the
pelagic environment, and will be key for predicting changes there induced, for example,
by climatic warming.
6.1 Introduction
The term pelagic means ‘of the open sea’ and the pelagic realm is a
largely open, unbounded environment in which the inhabitants have
freedom, within physiological limits, to move in three dimensions.
Contrary to the common perception of the sea as an unchanging,
relentless expanse, the open ocean is an environment where variability is
very much the norm. Patchiness in physical properties (e.g. temperature,
salinity, turbidity), biological production, and biomass exists at a range
of scales in space (centimetres to hundreds of kilometres) and time
(minutes to decades). One of the key challenges to understanding open-
ocean function lies in understanding the mechanisms that cause, and
consequences of, this patchiness (Mackas & Tsuda 1999).
Despite the fact that much of the open ocean is remote from land,
beyond the horizon for land-based observers, it has not escaped human
impacts. For example, 90% of stocks of large pelagic fish such as tuna
(Scombridae) and jacks (Carangidae) may have been removed by fishing
(Myers & Worm 2003), and whole zooplankton communities have
l The open ocean is an highly
variable environment.
shifted their spatial distribution (Beaugrand et al. 2002) possibly in
response to ocean warming, itself most likely caused by
anthropogenically-released greenhouse gasses (IPCC 2001).
The pioneering studies of open-ocean ecology made from vessels
with evocative names such as Discovery, Challenger, and Atlantis have
been enhanced in recent years with observations from technologically
advanced research platforms that include Earth-orbiting satellites
(Bricaud et al. 1999) and unmanned autonomous underwater vehicles
(AUVs) (Griffiths 2003). The aim of this chapter is to provide a synthesis
of open-ocean ecosystem function and the factors that control it, insight
into difficulties associated with sampling the heterogeneous pelagic
realm, and some examples of ecological step-changes (regime shift) in
the global pelagic environment.
6.2 Definitions and Environmental Features
The pelagic realm spans the entirety of the water column, beginning at
the sea surface and ending just above the seabed (the benthic realm;
Chapters 7 and 8). The pelagic realm can be subdivided by total water
depth and distance from shore. The neritic zone lies adjacent to shore,
over continental shelves, and covers about 8% of the Earth’s total sea
area. Out beyond the continental shelf break, which is delimited typic-
ally by the 200 m depth contour, lies the vast, open oceanic zone (92%
of the total sea area that covers 65% of the Earth’s surface). There are
numerous differences between the neritic and oceanic zones, differences
that arise not least because of differences in proximity to land and
consequent differences in nutrient and sediment loading in the water
column (much of the sediment load in the seas and oceans is terrigenous
and is delivered by rivers and estuaries; Chapter 4). Sailing out from a
coastal port towards the open sea it is common to notice a transition
from turbid to clear, blue waters. This transition is obvious not just from
the deck of a ship but is visible from space. Indeed, interpretations
of satellite remote sensed observations of ocean properties have to
distinguish ‘Case 1’ waters, where the ocean colour is determined pre-
dominantly by algal pigments, from ‘Case 2’ waters where reflections
from particulate matter dominate (Fig. 6.1) (Babin et al. 2003). One
adaptive biological consequence of the difference in sediment and par-
ticulate loading between the oceanic and neritic zones can perhaps be
seen in squid anatomy. Myopsin squid inhabit the neritic zone (for
example Loligo forbesi, which is common in north-west European
coastal waters, or L. opalescens from the west coast of the USA) and
have a membrane across the eye that may serve to protect the eye from
particulate irritants suspended in the water: in the open ocean myopsin
l Despite its apparent
remoteness, the open ocean has
been influenced strongly by
human activities such as fishing
and anthropogenic climate
warming.
l The neritic and oceanic zones
are very different, due in part to
their differing distances from the
coastline.
6.2 DEFINITIONS AND ENVIRONMENTAL FEATURES 213
squid are replaced largely by members of the suborder Oegopsina (for
example the European flying squid, Todarodes sagittatus) in which the
membrane is absent, possibly because it is unnecessary.
The pelagic component of the open ocean can be divided further by
depth. The upper surface of the ocean is known as the neustic zone and,
in the tropics especially, is an habitat made harsh by exposure to high
Case 1
Case 2
Fig. 6.1 An enhanced true-colour view of ocean colour from the SeaWiFS satellite
(18 May 1998 1308 GMT) showing Case 1 and 2 waters around and to the south-west
of the British Isles, North-West Europe. Satellite images were received by the NERC
Dundee Satellite Receiving Station and processed by Peter Miller and Gavin
Tilstone at the Plymouth Marine Laboratory (PML) Remote Sensing Group
(www.npm.ac.uk/rsdas/). SeaWiFS data courtesy of the NASA SeaWiFS project
and Orbital Sciences Corporation.
l The presence of a protective
eye membrane in neritic squid
may be an adaptation to
heavy particulate loading in
near-shore seas.
6: PELAGIC ECOSYSTEMS214
levels of ultraviolet radiation. Floating organisms inhabiting this zone
typically have a blue colouration (Fig. 6.2) due to the presence of pro-
tective pigments that are able to reflect this damaging part of the light
spectrum. The development of the ozone hole in the Earth’s atmosphere
has resulted in increased levels of UV radiation reaching the Earth’s
surface (the ozone layer acts as an UV shield), particularly in the
southern hemisphere. In the Southern Ocean, Antarctic krill (Euphausia
superba) may be particularly vulnerable to UV-induced DNA mutation
because krill DNA is rich in thymine, which is the base that is most
susceptible to UV radiation damage (Jarman et al. 1999). Since krill
migrate away from the sea surface during daylight hours, however, their
behaviour will probably serve to limit DNA damage, but UV damage to
other species remains a distinct possibility.
Light also plays an important role in pelagic ecosystem function away
from the neustic zone, both because it drives primary production
(Chapter 2) and because it enables visual predation (predators that hunt
using the sense of sight). In clear oceanic waters the threat from visual
predators is increased because these predators can detect prey over
greater ranges (Aksnes & Giske 1993). Shark attacks on humans often
occur in turbid waters (Cliff 1991), possibly because under these con-
ditions prey recognition is difficult and humans are mistaken for typical
prey such as seals. The upper part of the water column into which light
penetrates is called the photic zone. In clear tropical oceanic waters this
zone may extend as deep as 200 m (much less in more turbid, temperate
locations), although at this depth light intensity will usually be too low
to drive photosynthesis (Chapter 2).
The upper 200 m of the water column is also known as the epipelagic
(Fig. 6.3). Light at the red end of the spectrum is absorbed rapidly by
seawater and does not penetrate far into the epipelagic zone (Chapter 2).
Red colours are effectively invisible at depth, therefore, and many pelagic
crustaceans adopt this colour as a means of camouflage against visual
predators (Fig. 6.4). Below the epipelagic zone are, sequentially, the
mesopelagic (200 m to 2000 m), the bathypelagic (2000 m to 4000 m),
and the abyssopelagic (4000 m to 6000 m) zones. As depth increases,
organisms in the pelagic environment are faced with increasing physio-
logical challenges: pressure increases by 1 atmosphere for every 10 m
increase in depth (Box 6.1; 1 atmosphere¼ 1 kg cm�2 or approximately
the mass of a Mini on an area 25 cm� 25 cm), and in some locations
oxygen-minima layers (Rogers 2000) arise at depth because oxygen is
depleted by bacteria breaking down material sinking from the sea surface.
In general terms the total mass of biota per unit volume of seawater
decreases with depth (Yamaguchi et al. 2002). This is because, with
the exception of energy input by chemoautotrophic processes at
hydrothermal vents (Chapter 8), biological processes in the deep sea are
l Organisms that live in the
neustic zone are particularly
vulnerable to changes in UV
radiation.
l Daily and seasonal changes in
incident light intensity have
profound effects on biological
processes in the pelagic
environment, driving vertical
migrations and seasonal plankton
blooms.
Fig. 6.2 The Portuguese
man-o-war (Physalia physalis, a
colonial Cnidarian) floats
at the sea surface and has a
blue colouration. Photograph
from joaoguaresma.com with
permission.
l The total mass of biota per unit
volume of seawater decreases
with depth.
6.2 DEFINITIONS AND ENVIRONMENTAL FEATURES 215
fuelled entirely by photosynthetically–generated organic matter that
sinks from the illuminated surface region. As material sinks further from
the surface it becomes distributed through an ever-increasing volume of
water. This dilution effect means that as distance from the surface
increases, food availability decreases, with the consequence that less
animal biomass can be sustained at depth. In the deep pelagic zones,
Fig. 6.4 A mesopelagic
zooplankton/nekton sample
showing the predominance of
red-coloured crustaceans.
Photograph: Andrew Brierley.
Pelagic
NeriticOceanic
Epipelagic
Mesopelagic
200 m
1000 m
4000 m
6000 m
10000 m
Bathypelagic
Hadalpelagic
High water
Low water
Continental shelf
Benthic
Apho
tic
Abyssal
Hadal
Continental slopeBathyal
Littoral
Photic
Fig. 6.3 Depth zones.
6: PELAGIC ECOSYSTEMS216
energy efficiency is particularly important and animals adopt stealthy,
sedentary lifestyles and use cunning mechanisms to ambush prey in
darkness (Seibel et al. 2000).
6.3 Pelagic Inhabitants: Consequences of Size
Pelagic organisms can be divided in to two categories on the basis of
their locomotory prowess. Plankton are unable to counteract the influ-
ence of currents and drift passively in the horizontal plane (ocean cur-
rents often exceed 1 knot or c.0.5 m s�1). Plankton can move vertically,
adjusting their depth, and do so pronouncedly during diel vertical
migrations at dawn and dusk (Fig. 6.5). Diel vertical migrations are a
ubiquitous feature of pelagic ecosystems (Hays 2003) and are thought to
be driven primarily by the trade off required to enable plankton to feed
in the food-rich upper water column and yet to avoid the illuminated
upper layer in daylight because of the increased risk of visual predation
incurred there at that time (Tarling 2003). Some organisms, such as
copepods and chaetognaths, complete their entire life cycle as plankton
and are called holoplankton. Others, such as fish larvae and barnacle
Box 6.1 Seawater as a dense and viscous (sticky) medium
Seawater is denser and more viscous than air. The increased density offers advantages
to some marine organisms, for example providing physical support that reduces the need
for skeletal strength, but also presents challenges such as rapidly increasing pressure with
depth. Seawater density varies as a function of the concentration of dissolved salts
(salinity; salinity is reported without units since it is defined in terms of the ratio of the
electrical conductivity of seawater to the electrical conductivity of a potassium chloride
standard, see also Chapter 4) and temperature (warm water is less dense than cold
water). The strength of cohesion, or stickiness, of a fluid is quantified by its dynamic
viscosity. The dynamic viscosity of olive oil, for example, is 40 times that of seawater.
The motion of a particle of a given size and velocity is more impeded through a medium
of greater dynamic viscosity (think of marbles sinking through a bottle of water and a
bottle of olive oil). For a fluid of a given dynamic viscosity (e.g. seawater) the continuity of
motion of a particle is controlled by its velocity and size, and can be quantified by the
Reynolds number (Re). For a given fluid, Re is simply ((particle velocity�particle size)/
dynamic viscosity). Re is dimensionless because the units used in its calculation cancel
out. Broadly speaking for small organisms moving at slow speeds Re is less than 1000
and seawater is ‘sticky’: ciliates, for example, stop the moment they cease swimming. For
larger organisms travelling at higher speeds Re is greater than 1000: the inertia of a large
pelagic fish, for example, enables it to continue gliding through the water even after it
has ceased active swimming (see also Chapter 3).
l Energy efficiency is important
in the deep pelagic due to
the dispersed nature of
potential food.
l The word plankton is derived
from the Greek word planao,
which means ‘to wander’
6.3 PELAGIC INHABITANTS: CONSEQUENCES OF SIZE 217
larvae, spend only a part of their life cycle as plankton, either growing or
settling out to the seabed as they age, and are called meroplankton.
A planktic phase provides opportunities for dispersal and colonization,
but is also a stage that is particularly vulnerable to predation (Pechenik
1999, see also Chapter 13).
Organisms that are capable of swimming to the extent that they can
overcome currents are known as nekton. Inhabitants of the open ocean
span several orders of magnitude of size, ranging from viruses, bacteria,
and protozoa to large predators such as sharks and whales, which
may reach many metres in length and body masses of several tonnes.
Generally speaking, large organisms are nektic and smaller organisms
are planktic: micronekton have intermediate swimming abilities
and are of the order of 4 cm in length (for example large euphausiids).
This size-related difference in mobility arises in part because of the
interaction between size and viscosity (Box 6.1): seawater is essentially
a ‘sticky’ medium for small organisms (Van Duren & Videler 2003) and
a constant expenditure of energy is required to maintain their move-
ment. There are exceptions, however, and the Arctic lion’s mane jellyfish
(Cyanea arctica), for example, may attain tentacle lengths of 40 m
but is a passive, planktic drifter. The very smallest planktic organisms
include viruses and bacteria (Azam & Worden 2004) and protozoa
(Struder-Kypke & Montagnes 2002). The small size of these organisms
belies their importance to pelagic ecosystem function. Dissolved organic
carbon is taken up by bacteria, which are consumed by heterotrophic
nanoflagellates and in turn by ciliates in the so-called microbial loop at
the base of the food chain (Chapter 3). This loop recycles organic matter
l Animals whose larvae have a
planktic phase generally produce
a relatively high number of eggs
for their body mass.
100
200
300
400
500Dept
h (m
)
Longitude degrees
Echo
inten
sity,
dB
600
700
–38 –36 –34 –32 –30 –28 –26 –24
–50
–60
–70
–80
–90
Fig. 6.5 An echogram showing the rapid ascent of zooplankton and nekton (as
detected using a 38 kHz scientific echo-sounder) from depth to the near surface at
dusk. This migration from 400 m to 50 m takes less than 2 hours. The upper bar indicates
periods of day (light blue) and night (dark blue).
6: PELAGIC ECOSYSTEMS218
that is too small to be consumed by metazoan plankton, and the
metazoans are able to prey upon ciliates. The microbial loop therefore
fuels the pelagic food chain, and is especially important in oligotrophic
waters (Lenz 2000).
As well as impacting mobility, organism size is also a major architect
of pelagic food web structure. Pelagic organisms will typically consume
food items whole and the size of item that an animal can consume is
constrained by its mouth size such that predators are usually substan-
tially larger than their prey (Cohen et al. 1993; Jennings & Warr 2003).
This concept is well captured by Brueghel’s picture Big fish eat little fish,
in which small fish are tumbling out of the mouths of successively bigger
fish (Fig. 6.6). A more trophically extensive example from the North Sea
has unicellular algae such as diatoms (c.100 mm diameter), grazed by
copepods (e.g. Calanus finmarchicus, c.3 mm length), which are in turn
predated by herring (Clupea harengus, c.20 cm length), which might be
consumed by gannets (Sula bassana, wing span c.180 cm). This strongly
size-structured progression is in marked contrast to many terrestrial
food chains where small predators (for example hyenas, body mass
c.40 kg) may cooperate in social groups to take herbivores that are
considerably larger (e.g. wildebeest, c.250 kg).
Pelagic food webs are often far more complex than the simple four-
linked-chain diatom-copepod-fish-bird example from the North Sea. An
analysis of a 29-species food web for the Benguela ecosystem undertaken
l Very few pelagic organisms
dismantle their prey before
eating, a behaviour more common
in benthic biota such as crabs.
l In terrestrial systems, animals
that forage in social groups can
deal with prey items much larger
than themselves.
Fig. 6.6 ‘Big Fish Eat Little Fish:’
after, by Pieter Brueghel the
Elder (1557) showing smaller
fish tumbling from the mouths of
bigger fish. The Metropolitan
Museum of Art, Harris Brisbane
Dick Fund, 1917. (17.3.859)
6.3 PELAGIC INHABITANTS: CONSEQUENCES OF SIZE 219
to determine if culling Cape fur seals (Arctocephalus pusillus pusillus)
would increase hake (Merluccius spp.) biomass (Yodzis 2000) noted that
there were over 28 million pathways from hake to seals! This com-
plexity of food webs, and the fact that most levels can be controlled
either from above (top-down control, e.g. by predation) or below
(bottom-up control, e.g. by food limitation, Verity & Smetacek 1996)
renders it very difficult to make predictions of the consequences of
bioregulation. In general smaller mean predator : prey body size ratios
are characteristic of more stable environments, and food chains are
longer when mean predator : prey body size ratios are small (Jennings &
Warr 2003). Systems that have shorter food chains are generally much
more susceptible to trophic cascade effects (Chapter 7).
6.4 Temporal and Spatial Variabilityin Pelagic Ecosystems
The open ocean is not homogenous. Interactions between physical and
biological processes result in variability over a range of temporal and
spatial scales, and patchiness is a key feature of pelagic ecosystems.
The Stommel diagram (Fig. 6.7) illustrates the scales of variability that
are inherent characteristics of pelagic ecosystems, from centimetres to
thousands of kilometres and from seconds to millennia, and shows how
variations in time and space are interlinked. It is important to appreciate
the interplay of temporal and spatial scale, and it is a theme that recurs
throughout this book. Phytoplankton are short-lived and are influenced
by small-scale mixing processes (Martin 2003). The diel vertical
migration of zooplankton at dawn and dusk is restricted to certain times
of day but occurs everywhere (Pearre 2003). Fish, which are generally
longer lived than zooplankton, are impacted by environmental variab-
ility over longer time scales: the Peruvian anchoveta (Engraulis ringens),
for example, is influenced strongly by the El Nino Southern Oscillation
that exhibits decadal scale variability (Chavez et al. 2003, see also
Chapters 7, 12, and 14).
6.4.1 Physical processes contributing to temporaland spatial variability
The physical properties of the open ocean are heterogeneous by
depth, position (latitude, longitude), and over time. In addition to the
depth-related changes in light intensity and oxygen concentration already
mentioned, vertical gradients in water temperature, density, and nutrient
concentration may also exist. Solar heating warms the upper ocean
leading to the development of a thermal gradient by depth. In some
l The complexity of food webs
means it is difficult (if not
impossible) to predict the
outcome of selective culling of
higher predators on lower tropic
levels.
l Fish, which are generally
longer lived than zooplankton,
are impacted by environmental
variability over longer time scales.
6: PELAGIC ECOSYSTEMS220
l Where winds are light and
wave action slight, or in water
that is too deep to be mixed
completely by tidal flow, vertical
stratification can develop.
A
B
CD
F
H
G
E K
JI
9
6
3
02
46
810
121 cm
1 cm
100 cm
10 Km
10 Km
100 Km
1000 Km
10000 Km
SpaceTime
Ships
A. 'Micro' patchsB. SwarmsC. UpwellingD. Eddies amd ringsE. Island effectsF. 'El Nino' type eventsG. Small ocean basins'H. Biogeographical provincesI. Currents and oceanic fronts - lengthJ. CurrentsK. Oceanic fronts - width
Moorings
Satellites
Biomassvariability
Log L (cm)
Log P (sec)Minute
Hour
DayWeek
Month
Year
Century
1000 Y
10000 Y
Diel vertical migrationAnnual cycles
Ice age variations
Fig. 6.7 The Stommel diagram, overlain to show the scales that can be sampled with
various platforms, and features such as fronts. Modified from ICES Zooplankton
Methodology Manual, Harris et al. (eds., 2000), page 36, Copyright 2000, with
permission from Elsevier.
regions, combinations of tidal and wind-driven processes cause turbu-
lence and mixing of heated surface water with cooler water below.
However, in regions or seasons where winds are light and wave action
slight, or in water that is too deep to be mixed completely by tidal flow
(Chapter 7), pronounced vertical stratification can become established:
warm surface waters then become effectively isolated from cooler,
deeper waters by a thermocline. Vertical stratification is also promoted
in situations where fresh water is introduced. Rain, river run-off, and
ice-melt all introduce fresh water to the surface of the ocean. Low
salinity waters are less dense than high salinity waters (Box 6.1) and
stabilize the upper water column because more energy is required to mix
low salinity waters downwards. The strong density gradient between the
mixed, buoyant, low salinity surface waters and underlying high salinity
waters is known as the pycnocline. The depth of the mixed layer,
6.4 TEMPORAL AND SPATIAL VARIABILITY 221
as bounded by the pycnocline or thermocline, will vary depending upon
prevailing conditions (Chapters 2 and 7).
As well as causing downward mixing, wind can lead to the upward
transport of water from depth. Such wind-driven upwelling occurs over
a range of scales. At the small scale, Langmuir circulation is generated as
wind blows steadily across calm water, causing near-surface vortices
several metres in diameter to develop parallel to the wind flow (Box 6.2).
At the interfaces between neighbouring vortex cells, alternating lines of
upward and downward convergence develop. Flotsam accumulates on
the surface above the downward zones, leading to the development of
prominent, parallel wind lanes on the surface. Zooplankton may also
accumulate in downward zones because they are able to swim upwards
against the flow (Pershing et al. 2001).
At the large scale, wind plays a role inducing flow in most surface
currents. Currents do not flow parallel to the direction of the wind but,
due to interactions with the Coriolis force, when averaged over the whole
of the water column, currents move at 90� to the wind. Movement is to
Box 6.2 Wind-driven circulation processes
As wind blows over the surface of the sea it generates waves and induces vertical and
horizontal motion in the water. Langmuir cells are small-scale, parallel, helical vortices
that are often apparent as a series of wind lanes on the sea surface running parallel to
the direction of the wind. Vortices are usually not large enough to bring nutrients up
from deep water beneath the pycnocline.
Up welling(divergence)
50 m
Helical vortices
Down welling (convergences where seaweed, debris, foamand plankton accumulate)
Up
Down Down |continues
l Langmuir circulation can
lead to the accumulation of
zooplankton and flotsam, which
forms visible wind lanes, at the
sea surface.
6: PELAGIC ECOSYSTEMS222
the right of the wind in the northern hemisphere and to the left in the
south. This movement is called Ekman transport (Box 6.2). It con-
tributes significantly to general ocean circulation and can result in
pronounced upwelling. In the case of the Benguela Current, for example,
and other southern hemisphere eastern boundary currents, the prevailing
south-easterly winds blow alongshore and drive near-shore Ekman flow
away from the coast. This in turn draws cold, nutrient-rich waters from
depth to the surface at the coast (Carr & Kearns 2003). Changes in
global wind patterns have the potential to affect upwelling and the
ecosystems that are dependent upon them (Grantham et al. 2004).
Horizontal boundaries between water masses with different physical
properties are known as fronts (Box 6.3). Fronts occur at a range of
BOX 6.2 continued|
Fig. 1 Large-scale ocean currents are also induced by wind, but occur at angles to the
direction of the wind rather than parallel to it. This is because of the interaction between
friction and the Coriolis force. The Coriolis force is the force experienced by a moving
body of water due to the fact that the planet is rotating. The water column can be
thought of as a series of horizontal layers. The upper layer at the sea surface is subject to
wind friction (wind stress) at the top and water friction (eddy viscosity) at the bottom.
Subsequent layers are impacted by friction with layers above and beneath. Slippage
between layers result in an exponential decrease in current speed with depth until, below
the depth of frictional influence, wind influence ceases. Cumulative impacts of
Coriolis force result in an increasing angle of deviation away from the wind with depth.
Current vectors in all layers form a spiral pattern known as an Ekman spiral. The
averaged effect of the spiral is that the mean motion of the wind-driven (Ekman) layer is
at right angles to the wind direction. Reprinted from Ocean Circulation, 2nd ed., the
Open University Course Team 2001, pp. 42 and 68, Copyright 2001, with permission
from Elsevier.
Wind stress
Wind stress
Average motion of the Ekman layer
Average motion of the Ekman layer
Ekman layer
Depth offrictionalinfluence
Coriolis force
Coriolis force
6.4 TEMPORAL AND SPATIAL VARIABILITY 223
Box 6.3 Front formation and elevated biological activity at fronts
In the same way that fronts on weather maps mark boundaries between different air
masses, fronts in the sea are boundaries between dissimilar bodies of water. Fronts occur
off estuaries at boundaries between fresh and salt water, in shelf seas between mixed
and stratified waters, at continental shelf breaks adjacent to upwelling regions and, at
the global scale, between major current systems. Fronts tend to be sites with higher
biological activity that the surrounding water masses, often because nutrients are
transported upwards into the stratified euphotic zone at fronts.
Tidal mixing fronts (also known as shelf sea fronts) occur between tidally mixed and
stratified waters. They occur when the intensity of turbulent mixing caused by tidally
induced flow over the seabed is sufficient to overcome the barrier to mixing caused by
thermal stratification. In simple terms, this is a function of the strength of the tidal flow
and water depth; a strong tidal flow will generate sufficient turbulence to completely mix
shallow water. On the stratified side of the front, nutrient concentrations in the warm
surface waters are depleted and the strong thermal gradient prevents nutrients from
beneath being mixed upwards. Phytoplankton growth is therefore nutrient limited and
low. On the well-mixed side of the front, although nutrients are not limited, phyto-
plankton are continually mixed down out of the illuminated surface layer and growth is
light-limited. At the front itself stratification weakens sufficiently to enable some vertical
nutrient flux but remains strong enough to hold phytoplankton in the photic zone long
enough for them to take advantage of the nutrients. Increased phytoplankton produc-
tion at fronts leads to higher zooplankton standing stocks and increased densities of
predators and underlying benthos (see also Chapters 2 and 7).
Eddies
Fron
t
Warm stratified
Along-front flow
Thermocline
Weak nutrientflux to base ofthermocline onstratified side of front
Increased fluxthrough weakerstratified side at front
Shallow
Strong vertical exchange
Tidal flowcauses turbulance
Deep
6: PELAGIC ECOSYSTEMS224
scales, from tidal mixing fronts that separate mixed and stratified waters
in coastal seas (Hill et al. 1993) to major oceanographic boundaries such
as the North Wall of the Gulf Stream and the Antarctic Polar Front
(Taylor & Gangopadhyay 2001). High-velocity current jets associated
with fronts can be important long-distance transport routes for many
marine organisms: Antarctic krill (Euphausia superba) are transported
widely throughout the Scotia Sea from breeding centres off the Antarctic
Peninsula on frontal currents (Thorpe et al. 2004), and large oceanic
squid such as Illex illecebrosus take advantage of currents for distribu-
tion and feeding (O’Dor 1992). Meanders in fronts can lead to columns
of water (core rings) being shed from one side of the front to the other;
the retroflection of the Agulhus current around the southern tip of
Africa, for example, sheds warm core rings regularly into the south
Atlantic (Garzoli et al. 1999) and is an important mechanism for trans-
oceanic mixing.
6.4.2. Consequences of temporal and spatial physicalvariability for pelagic primary productivityand biogeography
The depth to which water column mixing occurs, the mixed layer depth,
has major implications for primary production in the open ocean
because photosynthesis only takes place in illuminated surface waters.
If the mixed layer is deep it is possible that phytoplankton will sink or be
carried down below the compensation depth (Chapter 2), reducing net
production. In temperate waters, primary production is minimal during
winter when light levels and temperatures are low and the upper water
column is thoroughly mixed by storm action and convection (Backhaus
et al. 2003). Phytoplankton blooms do not commence until calmer,
warmer weather in the spring leads to upper water column stratification.
From this point on, phytoplankton cells are retained in the upper mixed
layer, benefit from the increased illumination from the sun as it reaches
higher angles in the sky, and grow and reproduce rapidly. Phytoplankton
require nutrients such as phosphate, silicate, and nitrate to grow. As
phytoplankton blooms develop, concentrations of these nutrients
become depleted and the effective isolation of the mixed layer from the
larger nutrient pool beneath means that nutrients are not replenished. In
temperate waters, nutrient limitation may inhibit phytoplankton growth
throughout the summer months. A second bloom may though occur at
the onset of autumn when wind-driven mixing brings nutrient-rich
waters from beneath the pycnocline up in to the illuminated surface
layer (Diehl 2002).
In regions of the world where upwelling is persistent throughout the
year (Box 6.2) nutrients tend not to be limited and annual primary
l Most fronts are highly mobile,
dynamic features that cannot be
represented realistically by single,
static lines on charts.
l Although the development of
stratification is a necessary
precursor to bloom formation, the
persistence of an upper layer that
is effectively cut off by density
and temperature gradients from
the waters beneath can eventually
inhibit phytoplankton growth.
l A phytoplankton maximum
can develop at the pycnocline
late in the season as this is the
interface between nutrient-rich
deep water and illuminated
nutrient depleted surface
stratified waters.
6.4 TEMPORAL AND SPATIAL VARIABILITY 225
production levels are high. However, in El Nino years, changes in pre-
vailing weather conditions reduce the usually strong upwelling off the
coast of western South America and ensuing nutrient limitation has
dramatic negative consequences for primary production and fisheries in
coastal waters. Tropical ocean basins tend to be permanently stratified
and primary production levels are low. As a consequence surface waters
in these tropical regions lack particulate matter and are very clear: such
waters are termed oligotrophic. Other areas of the world ocean have low
phytoplankton biomass despite the presence of high nutrient con-
centrations. These HNLC (high nutrient, low chlorophyll) regions
include the Southern Ocean and Equatorial Pacific, and hypotheses
proposed to explain their existence include grazing pressure and absence
of trace elements, particularly iron (Chapter 2).
Global variation in the pattern of annual primary production is
strikingly clear in images of averaged chlorophyll concentration obtained
through satellite imagery (see Fig. 2.3, Chapter 2). Regional coherence
in the pattern of annual phytoplankton production has been used as one
diagnostic feature in the hierarchical separation of the global ocean into
distinct biogeographic biomes and provinces (Longhurst 1998). A biome
is the largest coherent community unit that it is convenient to recognize,
and Longhurst (1998) distinguishes four in the global ocean, which are
characterized by the principal mechanisms driving their mixed layer
depth: in the Westerlies biome local winds and irradiance force the
mixed layer depth; in the Trades biome the mixed layer depth is influ-
enced by large-scale ocean-circulation processes; in the Polar biome
the presence of buoyant, fresh water from ice melt in spring constrains
the mixed layer depth; and in the Coastal biome diverse processes
including upwelling force the mixed layer depth. Within these biomes,
51 provinces are recognized (Chapter 1). Separation of the global ocean
into provinces is very useful because it allows regional differences in
physical oceanography to be used to gain understanding, and make
predictions, of regional differences in ocean ecology.
6.4.3 Consequences for higher trophic levelsof variability in primary production
Regions of the world’s ocean with high primary productivity support
richer pelagic communities, with higher total biomass, than do regions
with low primary production. In fact there is a direct linear relationship
between the magnitude of annual primary production and nekton (fish
and squid) production (Sommer et al. 2002) (Fig. 6.8). This relationship
is apparent in the distribution of global fish catches (Chapter 12):
nutrient-rich shelf seas and regions with strong upwelling account for
the vast majority of the world’s commercial catch (Watson & Pauly
l Iron is thought to be one of the
key limiting trace elements in
HNLC regions, and experimental
iron fertilization in these
locations has stimulated
phytoplankton production
(Boyd 2002).
l Knowledge that a particular
area of ocean lies in a particular
province enables predictions to be
made regarding ecosystem
function there, even if field data
for the specific area are lacking.
6: PELAGIC ECOSYSTEMS226
2001), whereas oligotrophic central open-ocean basins contribute little.
Commercially important pelagic fish species do not consume phyto-
plankton directly but usually predate zooplankton and micronekton that
are primary consumers. Understanding zooplankton ecology is therefore
key to understanding fisheries production.
Zooplankton blooms are only able to develop once phytoplankton
biomass and production has become sufficient to sustain zooplankton
grazing rates. In temperate waters, therefore, peaks in zooplankton bio-
mass occur in spring and autumn slightly after the phytoplankton
blooms. In high latitudes, where seasonality is extreme and the phyto-
plankton bloom is limited to a single spring/summer peak, some zoo-
plankton species survive the dark, food impoverished winter months in
deep water in a dormant state called diapause. Copedods including
Calanus finmarchicus in the sub-Arctic north Atlantic and Calanoides
acutus in the Southern Ocean build up large stores of lipids during
summer feeding, a small proportion of which fuels their survival over
winter. At the end of summer, growth and development are arrested
and individuals sink, overwintering in a state of hibernation at depths
between 500 m and 2000 m (Box 6.4). In late winter or early spring
copepods emerge from diapause and migrate to the surface to spawn.
Because of the short production season, the timing of reproduction is
critical at high latitudes. By using lipid reserves accumulated in the
12 3
50 100 150 200 250 3000
5
10
15
20
25
4
5
6 7
8
910
Total phytoplankton production (g C m-2 yr-1)
Fish+
squid
prod
uctio
n (g w
et wt
m-2
yr-1
)
Fig. 6.8 The relationship between primary production and nekton production. Fish and
squid production¼ (0.095 Phytoplankton production) – 3.73, r2¼ 0.96. 1¼Atlantic
Ocean gyre centre, 2¼Atlantic ocean gyre boundaries, 3¼Hawaiian waters,
4¼Bothnian Sea, 5¼Gulf of Riga, 6¼Gulf of Finland, 7¼Baltic Sea, 8¼Nova Scotian
shelf, 9¼Gulf of Maine, 10¼Mid-Atlantic bight. Redrawn from Iverson 1990.
Copyright 2000 by the American Society of Limnology and Oceanography, Inc.
l GLOBEC (global ocean
ecosystem dynamics) http://
www.pml.ac.uk/globec is a
global research effort to
understand interactions between
primary consumers, higher trophic
levels and fisheries.
l Calanus finmarchicus survives
food depleted winter months in
the north Atlantic by drawing on
lipid reserves. This also enables
them to produce young in
advance of the phytoplankton
bloom in the following year.
6.4 TEMPORAL AND SPATIAL VARIABILITY 227
previous year to fuel reproduction, these copepods can spawn early,
independent of the present year’s phytoplankton bloom, and ensure that
their young are in place to fully exploit the short livid phytoplankton
bloom.
The timing of the phytoplankton bloom not only influences zoo-
plankton secondary production but also places a significant control
Box 6.4 Diapause depth, water density, and lipid composition
The overwintering depth of the copepod Calanus finmarchicus varies throughout its
distribution range in the north Atlantic. In the eastern Norwegian Sea, for example,
Calanus overwinters at about 800 m whereas in the Iceland Basin overwintering is at
around 1500 m (Heath et al. 2004), despite the fact that both locations have similar total
water depths (c.2000 m). The water column vertical temperature profile varies markedly
throughout the north Atlantic: in the eastern Norwegian Sea the temperature at the
overwintering depth is approximately 0 �C whereas in the Iceland Basin it is much
warmer (4 �C). This physical variation provides much insight into the variation of the
overwintering depth. At the onset of diapause Calanus becomes physically inactive and
sinks passively until it reaches the depth where it is neutrally buoyant. This is the depth at
which the density of the copepod is the same as the density of the surrounding seawater,
which itself is a function of ambient temperature and salinity. One of the major con-
tributors to variation in density between copepods is lipid composition, such that density
decreases as the proportions of lipid increases. Using knowledge of the temperature and
salinity at the overwintering depth in several locations, and hence the density there, it has
been possible to predict the proportion of lipid that should be expected in individuals
overwintering at particular locations. A very good linear relationship has been found
between predicted and actual values (Heath et al. 2004).
It is not yet clear what the main evolutionary drivers of variation in copepod lipid
composition are. In order to survive the winter individuals need to descend below the
depth of winter mixing, and will also have increased chances of survival overwinter if they
descend below depths where predators can operate. In Norwegian fjords overwintering
is shallower in fjords where visual predators are absent (Bagoien et al. 2001), but the
winter distribution of predators in the open ocean is not yet known. Ironically, therefore,
individuals that feed too successfully over summer and lay down excessive lipid reserves
may be unable to sink to depths below predators; this would provide a strong selective
pressure against over consumption and may be one reason why some copepods enter
diapause early in the year when the phytoplankton bloom is still in full swing. Since
current flow is different at different depths, the overwintering depth will also have a
profound influence on the location at which individual Calanus surface after diapuse.
In order to complete its life cycle successfully Calanus has to surface at a location where
offspring can be spawned and hatch and eat, and descend to overwinter to complete the
life cycle. It is likely that few depths will enable this to be achieved, providing another
source of selection. It is clear that the life cycle of Calanus is tied very closely to its
environment; understanding the environmental space-time dynamics will be vital to
gaining full understanding of regional variability in Calanus abundance and con-
sequences to higher trophic levels such as fisheries.
6: PELAGIC ECOSYSTEMS228
upon organisms that predate upon zooplankton. Developing fish larvae
‘surf on the wave of zooplankton production’ and the match-mismatch
hypothesis (Cushing 1990) proposes that larval fish survival will be
greatest in years when the period of plankton production overlaps
most closely with the period of larval food demand. Satellite data
that revealed between-year variations in the timing of the spring
phytoplankton bloom support this hypothesis with regard to haddock
(Melanogrammus aeglefinus) on the shelf east of Nova Scotia (Platt et al.
2003; Fig. 6.9). An index of larval haddock survival (size of age 1 year
class divided by spawning stock biomass) showed that two exceptionally
strong year classes occurred when the peak of the spring phytoplankton
bloom was between 2 and 3 weeks earlier than the long-term average.
An early bloom may result in lower larval mortality caused by starvation.
Spatial as well as temporal coherence is required between production
and consumption if high-biomass pelagic communities are to develop.
At the large scale, it has been suggested that iron fertilization in regions
of low primary productivity could enhance production up the food chain
to zooplankton and fisheries, and that the increased photosynthesis
could also lead to an increased drawdown of atmospheric carbon
dioxide that may mitigate against climate change. Complex trophic
interactions, however, make implementation of this far from straight-
forward (Buesseler & Boyd 2003; Gnanadesikan et al. 2003). At the
smaller scale, patchiness is essential for maintaining ecosystem function.
If phytoplankton and grazers were mixed homogenously then resource
depletion would soon occur, whereas spatial segregation enables higher
l Survival of larval fish is greatest
when the overlap between the
period of planktic food
availability and food demand by
fish is greatest.
2
4
6
8
10
2 1 0 -1 -2 -3
Anomalies in the timing of spring blooms (weeks)
Survi
val in
dex (
R/SS
B)
20011980
20001979, 1997
1998
1981
1999
Fig. 6.9 Larval haddock survival (size of the year class at age 1 (R) divided by spawning
stock biomass (SSB)) against deviation from the mean time of the annual peak in
phytoplankton production. Timing of the peak of the spring phytoplankton bloom
explains 89% of the variance in larval survival, providing strong support for the match-
mismatch hypothesis. Redrawn from Platt et al. 2003 with permission from the author.
6.4 TEMPORAL AND SPATIAL VARIABILITY 229
overall biomass to be maintained (Brentnall et al. 2003). Furthermore,
if zooplankton did not aggregate in high densities then pelagic filter
feeders such as basking sharks and baleen whales would not be able to
survive on a diet of these organisms. Whales incur high energetic costs
while foraging (Acevedo-Gutierrez et al. 2002) and it has been esti-
mated, for example, that Right whales (Eubalaena glacialis) require
copepod prey concentrations to exceed a minimum threshold of 4500
individuals m�3 of seawater just to balance the energy expended during
feeding (Beardsley et al. 1996).
Frontal regions tend to be characterized by increased primary pro-
duction and to support particularly rich pelagic communities (Box 6.3).
Fronts are therefore sites of intense feeding activity and are targeted by
mobile predators including fish, squid, marine mammals, and birds
(Durazo et al. 1998). The exact mechanisms by which predators locate
prey in the wide expanse of the open ocean remain largely unknown and
different cues are likely to be important at different spatial scales
(Fauchald et al. 2000). In some ecosystems prey are located regularly in
production ‘hot spots’ (Davoren et al. 2003), attracting larger predators
in a predictable manner. Conservation measures aimed at reducing
conflict between wildlife and fishers need to account for geographic
variability such as this in ecosystem management plans. Fisheries for
Antarctic krill, for example, may in future be required to operate outside
the foraging areas of land-based central-place foragers during their
breeding season (Constable & Nicol 2002).
6.5 Sampling the Open Ocean
The sometimes-extreme horizontal, vertical, and temporal patchiness
that is characteristic of pelagic ecosystems, and the huge size range of
organisms inhabiting the open ocean, present considerable difficulties
for quantitative sampling.
Early studies of the open ocean depended almost completely on nets to
sample living organisms. Netting remains an important component of
biological oceanographic research, but the systems in use today are
considerably more complex than those used in the pioneering days
(Wiebe & Benfield 2003). These days nets are often equipped with depth,
temperature, salinity, and other sensors. Data from these sensors can be
relayed to the ship in real time, either along conducting cables or via
acoustic links, and enable nets to be placed accurately in the section of the
water column of particular interest (Brierley et al. 1998). Optical particle
counters (OPCs) can be used instead of nets to obtain estimates of
zooplankton numerical density (Heath 1995), and photographic and video
devices allow high-quality images of ocean inhabitants to be obtained
l If phytoplankton and grazers
were mixed homogenously then
resource depletion would soon
occur, whereas spatial
segregation enables higher
overall biomass to be maintained.
l Conservation measures aimed
at reducing conflict between
wildlife and fishers need to take
in to account the fact that
predators forage at production
hotspots that occur at
oceanographic features such
as fronts.
6: PELAGIC ECOSYSTEMS230
(Benfield et al. 1996). Light is attenuated rapidly by seawater though, and
visual sampling is often constrained by water clarity (Chapter 2). Sound,
on the other hand, propagates very efficiently through seawater, as is
testified by the long-range vocal communications of some whales.
Scientific echo-sounders can be used to detect and quantify abundance of
zooplankton and fish (Holliday & Pieper 1995), and ever-increasing
sampling resolution is detecting biologically important features such as
micro-layers (McManus et al. 2003), which are likely to be of very major
importance to pelagic ecosystem function (Box 6.5).
At the larger scale, organisms that must come to the sea surface to
breath (e.g. seals, whales), or that forage over the sea surface (e.g.
seabirds) can be counted at sea by observers on research vessels.
Although much biological oceanographic data is still collected from
ships, logistic constraints place restrictions on the amount of time that
ships can spend at sea. In order to make longer-term observations, or
observations over large extents of ocean, other sampling platforms or
techniques are required. Moored instruments can be used to collect long
time-series of data from spot locations (Schofield et al. 2002). Auto-
nomous underwater vehicles (Box 6.6) have the potential to be able to
operate in weather conditions that curtail sampling from ships, and in
addition can work in environments that are impenetrable to ships.
Earth-orbiting satellites are able to provide coverage of the entire surface
of the global ocean on a weekly basis and deliver near-synoptic infor-
mation on, for example, sea surface temperature, chlorophyll concen-
tration, and frontal position (Miller 2004).
Satellites can also be used to track the movements of larger animals as
they forage at sea over extended periods of time (Thompson et al. 2003).
Box 6.5 Microlayers
Advances in the resolution of optical and acoustic sampling technology have lead to the
discovery of widespread ‘thin layers’ of high biological activity in the ocean. These layers,
which range in thickness from a few centimetres to a few metres, may be many kilo-
metres in horizontal extent and may persist for several days. They contain densities of
organisms several orders of magnitude higher than adjacent depth zones, and layers at
different depths in the same area may contain distinct plankton assemblages. Thin layers
produce microenvironments of physical, chemical, and biological parameters. Microlayers
sometimes occur at the pycnocline as zooplankton forage on material that is suspended
there, but may be deeper or shallower. Layers tend to occur in stratified water where
current sheer is low. The species or populations that comprise each distinct thin layer
probably aggregate in response to different sets of biological and/or physical processes.
The existence and persistence of planktic thin layers generates great biological hetero-
geneity in the water column, and may go some way to explain the ‘paradox of the
plankton’ in which high species diversity occurs in small, apparently homogeneous bodies
of water.
l While modern imaging and
acoustic technology has improved
our ability to sample the ocean
realm, we continue to be reliant
upon often-rudimentary nets to
obtain biological samples.
l Satellites can relay information
from tags attached to air
breathing animals such as whales
and turtles, collecting data about
their behaviour and patterns of
movement in real time.
6.5 SAMPLING THE OPEN OCEAN 231
Leatherback turtles (Dermochelys coriacea), for example, have been
tracked in the north Atlantic (Hays et al. 2004). Leatherbacks are crit-
ically endangered, and a major source of mortality for them is capture
by pelagic fisheries. Knowledge of Leatherback distribution and dive
characteristics obtained via satellite telemetry could lead to the imple-
mentation of conservation measures designed to reduce the interaction
of turtles with fisheries, and thus reduce by-catch.
The capacity of scientists to be able to collect data from the pelagic
realm seems ever to be increasing. Plans are afoot to establish a series of
permanent, automated ocean observatories that will be able to deliver
multidisciplinary data continuously in real time, year on year. Although
these systems will contribute enormously to our understanding of
ocean ecosystem function, they will present new challenges in terms
of extracting meaningful summaries from potentially overwhelming
quantities of data.
6.6 Pelagic Fisheries
Fisheries for pelagic species have the potential to be among the most
sustainable and least damaging to the environment. Shoaling species like
Box 6.6 Autonomous Underwater Vehicles
Autonomous Underwater Vehicles (AUVs) are unmanned submersibles that can be
programmed to navigate in three dimensions underwater. They can carry a variety of
scientific instruments and are able to make measurements in parts of the ocean that are
inaccessible, either physically or operationally, to conventional research platforms such as
ships. The Autosub AUV, for example, has been
equipped with a scientific echo-sounder and
deployed on missions beneath Antarctic sea ice.
There, it has made observations on the distribution
of krill under ice, and of ice thickness, that were
impossible to make using ice-breaking research
vessels (Brierley et al. 2002). Autosub is among the
largest of AUVs presently available to the scientific
community, with an instrument payload capacity of
100 kg (weight in water). Autosub is 7 m long�1 m
in diameter, weighs 2400 kg, is powered by man-
ganese alkali batteries, propeller-driven, has a range
of about 800 km, and a maximum depth capability
of 1600 m. Web link: http://www.soc.soton.ac.uk/
OED/index.php?page¼ as
6: PELAGIC ECOSYSTEMS232
the Atlantic mackerel (Scomber scombrus) and North Sea herring
(Clupea harengus) form single-species aggregations and by-catch is
minimal (Chapter 12). Indeed, at the time of writing, stocks of these two
species seem to be bucking the global trend of decline and are thriving
under good management and regulation. Myctophids, or lantern fish,
are small mesopelagic fish that form a major component of oceanic deep
scattering layers. They have been fished historically in the south-west
Indian Ocean and in the south Atlantic, but fishing ceased in 1992
because of unfavourable economics and market-resistance, and mycto-
phids are not presently under threat.
Planktivorous forage fish such as sardine and anchovy have vital eco-
system functions, particularly in upwelling zones, where they typify
mid-trophic-level wasp-waist ecosystems (Cury et al. 2000). Abundance
of these species can fluctuate wildly under variable environmental regimes
and high fishing pressure and may result in major ecosystem changes.
Fishing for large tropical pelagic fish including tuna and jacks has also
had substantial impact. Analyses of long-lining data suggest that 90% of
biomass of large pelagic fish may have been removed (Myers & Worm
2003). Open-ocean fisheries have tended to develop ahead of manage-
ment procedures (maybe by as much as 15 years) and, in the case of
pelagic long-line fisheries, it is possible that the estimated pre-exploitation
biomass to which management processes are anchored are unrealistically
low because they represent already-depleted stock levels. This ‘missing
baseline’ presents particular difficulty for the long-term restoration of
stocks because the size of the stock pre-exploitation remains unknown,
and thus it is very difficult to say when or if restoration has been achieved.
Not only has long-lining hit pelagic fish hard, but it has been and is still
responsible for substantial declines in albatross populations. Albatrosses
take baited long-line hooks as they are thrown from fishing vessels,
drowning as the long-line sinks, and some populations are showing
marked and continuing decline (Tuck et al. 2001). Seine netting for tuna
also suffers from by-catch, particularly of dolphins, although, following
international outcry, practices are in place to reduce the impact to levels
that are now ecologically sustainable (Hall 1998).
Squid are fished on the open seas, mostly using hooked, coloured lures
(jigs) at night to catch animals attracted to bright lights. The high
intensity lights used to attract squid to jigs are so bright that they can be
seen from satellites, and this has opened a new mechanism for poten-
tially monitoring and managing open-ocean squid fisheries (Rodhouse
et al. 2001). Robust management is particularly important for squid
because they are short lived, often semelparous (spawn once and die),
species and therefore vulnerable to over-fishing since there are few
cohorts to provide a buffer from failure of any single generation. Squid
also respond rapidly to changing oceanographic conditions and it is
l Fisheries for pelagic species
have the potential to be among
the most sustainable and least
damaging to the environment.
l It is possible that the estimated
pre-exploitation pelagic fish
biomass to which management
processes are anchored are
unrealistically low because they
represent already-depleted
stock levels, the so called
‘missing-baseline’ effect.
l The whole issue of
‘dolphin-safe’ tuna remains the
subject of debate and there are
moves in the United States to ease
legislation relating to the
definition of the product.
6.6 PELAGIC FISHERIES 233
becoming increasingly clear that it is essential to understand interactions
between squid and their ocean environment in order to predict inter-
annual variations in recruitment. Recruitment of the squid Illex argen-
tinus to the Falkland Islands fishery, for example, increases in years
when water temperatures over the squid egg-hatching grounds are
favourable (16 to 18 �C). In years when movement of the highly
dynamic front between the Brazil and Falkland currents displaces waters
of favourable temperature from over the hatching area, recruitment is
reduced (Waluda et al. 2001).
6.7 Regime Shifts in Pelagic Marine Ecosystems
Ecologists have long recognized that ecosystems may exist in ‘multiple
stable states’. In the oceans, conspicuous jumps from one state to
another have become known as regime shifts (Scheffer et al. 2001).
Shifts typically take less than one year to occur and regimes may persist
for decades (Hare & Mantua 2000). Regime shifts may be driven by
climatic changes, fishing pressure, or both, and may be manifest in
parameters that measure physical and biological ecosystem state. In the
north Pacific, statistically significant regime shifts in 1977 and 1989 are
apparent in a composite index of 100 biological and physical time series
including the Pacific Decadal Oscillation (PDO), zooplankton biomass
estimates and salmon catches (Fig. 6.10).
Regime shifts present major challenges for scientists attempting to
manage fisheries. In the North Sea a regime shift in 1988 was evident
from plankton time-series data from the Continuous Plankton Recorder
(CPR) surveys (Reid et al. 2001). It has been suggested that this shift
was caused by increasing flow of Atlantic waters into the North Sea,
an increase that was correlated with a change in the North Atlantic
Oscillation Index (NAOI). Recruitment of cod (Gadus morhua) in the
North Sea has declined since the mid 1980s and it is possible that
changes in the plankton following the regime shift have had a negative
impact on the supply of food to young cod (bottom-up control)
(Beaugrand et al. 2003). In the face of such possible environmental
impacts on fisheries, it is clear that future attempts to manage fisheries
will need to take environmental factors into account as well as data on
fish population dynamics and catch levels. This realization has led to
calls for the development of an holistic, ecosystem approach to fisheries
management (Pitkitch et al. 2004; Chapters 12 and 15).
Although not necessarily a symptom of regime shift per se, jellyfish
appear to have increased in prominence in many pelagic marine eco-
systems worldwide in recent years (Mills 2001). Jellyfish blooms have
occurred in the Bering Sea, the northern Benguela current and elsewhere,
l Semelparous animals such as
squid are particularly vulnerable
to over-fishing as their
populations are not composed
of multiple cohorts that
provide a safety net against
over-exploitation.
l The 1988 regime shift in the
North Sea may have been caused
by a change in the North Atlantic
Oscillation.
6: PELAGIC ECOSYSTEMS234
possibly in response to climate and fishing effects. Indeed, it has been
suggested that jellyfish-dominated communities are the inevitable end
point in pelagic ecosystems perturbed by fishing (Pauly & MacLean 2003).
In the North Sea, correlations between the abundance of jellyfish and an
Index describing the periodically fluctuating North Atlantic Oscillation
(NAOI) have been detected (Lynam et al. 2004). Furthermore it seems
as though the recruitment of herring (Clupea harengus) in the North Sea
is adversely effected by high jellyfish abundance (although it is not clear
as yet whether this is due to predation by jellyfish on herring eggs or
larvae, competition between jellyfish and herring for zooplankton
food, or both). Complex interactions between climate and jellyfish may
therefore impact fish stocks, even in the absence of fishing, and could
have major implications for the recovery of fish stocks even after any
cessation of fishing.
1977 regime shift
Ecos
ystem
stat
e
1965 1975 1980 1985 19901965
0.5
1.0
0
-0.5
-1.0
1989 regime shift
Ecos
ystem
stat
e
1980 1985 1990 1995 20001975
0.5
1.0
0
-0.5
-1.0
Fig. 6.10 Mean and standard error of a composite index of 31 physical and 69 biological
parameters from the north Pacific between 1965 and 1997, showing significant step
changes or ‘regime shifts’ in 1977 and 1989. The physical time series represent
atmospheric and oceanic processes, while the biological time series all relate to oceanic
species ranging from zooplankton to salmon and groundfish. Each of the time series was
normalized before plotting and statistical analysis by subtracting the mean across both
regimes and then dividing the data for each regime by the standard deviation for that
regime. Standard errors for each year were computed as s/ffiffiffi
np
, where s is the standard
deviation across all variables within a year and n is the number of time series used in the
calculation (�100). Redrawn from Scheffer et al. 2001 with Nature Publishing Group’s
copyright permission and permission from the author.
l Jellyfish-dominated pelagic
communities may be one
consequence of overexploitation
of pelagic fish stocks.
l The NAO is presently in a high
phase, possibly restricting jellyfish
abundance. If the NAO were to
switch phase, climatic inhibition
of jellyfish abundance may be
relaxed and their numbers may
increase with adverse
consequences for fish stocks.
6.7 REGIME SHIFTS IN PELAGIC MARINE ECOSYSTEMS 235
There has been an almost exponential rise in the incidence of the
term ‘regime shift’ in the scientific literature since the early 1990s. It is
possible that this is because the incidence of regime shifts is increasing,
or that accumulating time-series of data are enabling more changes to be
detected. A note of caution is perhaps necessary regarding this apparent
increasing prevalence however: simulation studies looking at random,
independent time series with the same frequency content as the Pacific
Decadal Oscillation have shown that techniques used to identify ‘regime
shifts’ may find them in noise. Detection of step-changes does not
therefore necessarily provide evidence of processes leading to any
meaningful regime shift (Rudnick & Davis 2003) since the step changes
may be artefacts of the data.
6.8 The Future for Pelagic Marine Ecosystems
With an ever-increasing human population, and an ever-growing
demand for food protein, it seems likely that pressure on the open
ocean is likely to continue to grow. There is a history of fisheries
advancing further from shore, into deeper and more distant waters, as
conventional coastal resources are depleted and this looks set to con-
tinue. Fishing effort has already had major impacts on the global ocean.
As traditional fish species are removed, fishing effort turns from these
higher-trophic-level predators to smaller species. This phenomenon has
become known as fishing down the food web (Pauly et al. 1998) and is
ecologically unsustainable.
Humans are not just altering the open-ocean ecosystems by removing
biomass but are also degrading it by addition. The incidence of waste in
the ocean is increasing, with floating rubbish potentially distributing
species far beyond their usual ranges, leading to alien colonizations of
distant shores (Barnes 2002). Introductions of alien species in ballast
water from cargo ships has also had devastating effects on pelagic
ecosystems, such as the introduction of the ctenophore Mnemiopsis
leydii to the Black Sea (Kideys 2002). This ctenophore, a native of the
eastern USA, was predatory upon fish eggs and led to the collapse of the
Black Sea anchovy fishery. Dumping CO2 at sea in an attempt to reduce
further increases in atmospheric concentrations is being investigated
(Hunter 1999). As well as the addition of objects and organisms, human
activity has also increased noise levels in the ocean. Low frequency noise
from shipping, oil-exploration, and military activities may adversely
impact cetacean communication and foraging (Croll et al. 2001) by
masking the sounds these animals generate. Killer whales (Orcinus orca)
in the waters of Washington State, USA, increase the lengths of
their calls significantly (by about 15%) in the presence of whale-watcher
l The incidence of waste in the
ocean is increasing, with floating
rubbish potentially distributing
species far beyond their usual
ranges leading to alien
colonizations of distant shores.
6: PELAGIC ECOSYSTEMS236
boat traffic, and probably do so in an attempt to overcome the noise
generated by these boats that may mask their usual calls (Foote et al.
2004).
A recent forecast of the likely state of aquatic ecosystems in 2025
identified climate warming as the most significant single threat (Chapter
14), and climate changes have already had measurable impacts on sea
ice extent and zooplankton distributions (Polunin, 2005). Perhaps the
biggest climate-related threat to pelagic marine ecosystems arises
from the possibility that increased warming and consequent freshening
of the Arctic may switch off the north Atlantic current and hence per-
turb global ocean circulation (Rahmstorf 2002). It is probable that
changes like this have happened multiple times during the Earth’s his-
tory, and occurred over very short periods. If, as some models predict,
this were to happen again in the near future, the consequences for
the Earth’s ecosystem and climate would be so severe that concern
for the state of the pelagic realm would probably not be at the top of
humanity’s agenda.
l CHAPTER SUMMARY
� The pelagic realm is highly heterogeneous, and production is patchy in both space and
time. Generally production is higher closer to land, because of increased nutrient input
(rivers, upwelling), and close to the surface because of light availability. There is a
direct link between primary production and fisheries production.
� Organism size has a major bearing on mobility in the pelagic environment. Plankton
are generally small (<10 mm long) and are unable to swim against currents and drift
passively on them. Larger organisms (nekton) can move actively against currents.
Plankton can however move vertically and undertake pronounced diel migrations.
� Pelagic food webs are size-structured: small organisms are consumed by a succession
of larger grazers or predators. Most biomass occurs at the lowest trophic levels
(grazers) and gradually decreases at increasingly higher trophic levels.
� Environmental heterogeneity and the large range of pelagic organism-size (from
plankton to whales) presents a severe challenge for sampling the pelagic environ-
ment. Technological advances provide the means to collect ever-increasing quantities
of data, but net sampling remains important for collection of biological material.
� Pelagic fish that form large single-species shoals should be amongst the most
straightforward to manage and can be exploited with little risk of bycatch. Never-
theless, even pelagic species that inhabit remote locations far from land have been
impacted severely by fishing.
� Pelagic ecosystems can suffer step-changes, shifting rapidly from one state to
another. Such regime shifts may be due to impacts of climatic change, and have major
implications for ecosystem management.
CHAPTER SUMMARY 237
l FURTHER READING
Longhurst (1998) provides an excellent description of the causes and consequences of
geographic variability throughout the world’s ocean. Mann and Lazier (1998) give a
broad coverage of biological responses to physical processes in the ocean. A useful
plankton atlas of the North Atlantic was published in Volume 278 of Marine Ecology
Progress Series (2004). This provides a summary of Continuous Plankton Recorder (CPR)
methods, and describes how this invaluable long-term record has become an important
implement in our understanding of how pelagic ecosystems respond to global change.
Steele (2004) provides a brief review of regime shifts and their definition. His article is
the first article in a special issue dedicated to regime shifts.
� Anonymous 2004. Continuous plankton records: Plankton atlas of the North Atlantic
Ocean (1958–1999). Marine Ecology Progress Series 278, Supplement Available
on line at http://www.int-res.com/abstracts/meps/CPRatlas/contents.html
� Longhurst, A. R. 1998. Ecological Geography of the Sea. Academic Press, San Diego.
� Mann, K. H. & Lazier, J. R. N. 1996. Dynamics of Marine Ecosystems:
Biological-Physical Interactions in the Oceans. Blackwell, Oxford.
� Steele, J. H. 2004. Regime shifts in the ocean: reconciling observations and
theory. Progress in Oceanography 60: 135–41.
6: PELAGIC ECOSYSTEMS238
